Model Context Protocol Server for NebulaGraph

What is Model Context Protocol Server for NebulaGraph?

This MCP server provides seamless access to NebulaGraph 3.x, enabling LLM tooling systems to perform graph exploration, schema queries, and shortcut algorithms through the Model Context Protocol. It is designed for developers and AI agents who need to interact with NebulaGraph databases via natural language or LLM-driven workflows.

How to use Model Context Protocol Server for NebulaGraph?

Install via pip install nebulagraph-mcp-server. Configure connection by setting environment variables or using a .env file with NEBULA_VERSION=v3, NEBULA_HOST, NEBULA_PORT, NEBULA_USER, and NEBULA_PASSWORD. Then run the server with the nebulagraph-mcp-server command. For development, use npx @modelcontextprotocol/inspector uv run nebulagraph-mcp-server.

Key features of Model Context Protocol Server for NebulaGraph

• Direct access to NebulaGraph 3.x.
• Supports schema exploration and graph querying.
• Includes a few shortcut algorithms.
• Follows the Model Context Protocol for LLM integration.
• Simple CLI with environment variable and .env file support.

Use cases of Model Context Protocol Server for NebulaGraph

• Connect an LLM agent to a NebulaGraph database for natural language graph queries.
• Explore graph schemas and run predefined algorithms without writing low-level code.
• Power AI‑driven graph analysis tools that require real‑time graph data access.

FAQ from Model Context Protocol Server for NebulaGraph

Which version of NebulaGraph is supported?

Only NebulaGraph v3 is currently supported. The NEBULA_VERSION environment variable must be set to v3.

What are the runtime requirements?

Python 3.x and the nebulagraph-mcp-server package (installed via pip). A running NebulaGraph server accessible at the configured host and port is required.

How do I configure authentication?

Provide the NebulaGraph username and password via the NEBULA_USER and NEBULA_PASSWORD environment variables or in the .env file.

Does this server support MCP transports besides stdio?

The README does not specify alternative transports. It is designed to follow the Model Context Protocol, and the examples use the inspector tool which typically uses stdio.

Where does the graph data reside?

All data remains in the connected NebulaGraph database. This server only provides query and exploration capabilities; it does not store data locally.